Linger

83 hours to quantify a sense

FigBuild 2026 asked us to quantify a human sense. We picked smell.

We were a team of 4, and 2 of us had never partaken in a designathon. As we brainstormed, we decided that smell was not only intriguing, but also fun to speculate about.

I pushed, consistently, for ideas that could benefit someone in a serious way. For smell, that meant Alzheimer's and dementia: accessing memories through a sense, on command.

Two ideas we scrapped first

Before scent, we had two ambitious concepts on the table.

• Biophilic AR. A speculative space-station UI that would swap sterile surroundings for responsive natural scenery, calibrated to your biophilic response. Scope consumed too many resources. AR was a skill gap, and the problem space was too open.
• A sensory-deprivation cloak. An app-controlled wearable that compresses your sense of time, built around pain and boredom: MRIs for chronic-pain patients, and long boring flights. The team didn't align, and the vision of our app prototype felt lackluster.

We returned to scent. It was feasible in 83 hours, and we had a shared visual excitement about what scent could look like. The use case shifted early: our first instinct was a scanner for identifying someone's perfume, but it didn't feel weighty or useful enough. A niche idea like that was good, but we wanted to use this opportunity to think outside the box. Once we reframed around helping someone with Alzheimer's re-access a memory, the project earned its scope.

A scent app, and the wearable it would need to exist

The app does four things: detect, archive, map, and emit (we consider detection and archiving to be part of 1 feature). A speculative wearable then entered the picture, because we had to emit from somewhere. A brain chip felt too easy (as it could be argued for any project in this designathon). A nose-worn attachment was the most honest delivery mechanism for our concept, it might seem basic, but we wanted it to feel like a product you could easily imagine. Most of our time, though, was spent on the app.

Orbs as a diegetic interface

Before orbs, we tried perfume-bottles, gradients, and a few other directions. The orbs just felt right to all four of us. We all felt like we knew what they were attempting to portray, through speculation. They stayed customizable in color, they felt abstract enough to hold meaning without prescribing it, and they doubled as a living home screen where favorite scents grow with each scan.

• Vibrancy = scent intensity
• Hue = chemical family
• Diffusion = scent longevity and range

Demo

Interactive Prototype

Select a flow to explore below.

Try interactive prototype

Loading prototype…

Open in Figma

Designing around responsibility

Olfactory data felt deeply personal. When I think of my favorite scent, I feel as if it's a memory that belongs only to me. The brief pushed us to heavily consider this. Four safeguards built into the product:

• Ghost zones: user-defined areas where scent detection is automatically turned off, so others around you aren't quietly being scanned.
• Cognitive-overload prevention: the app caps scent intensity and how many signals surface at once, so users aren't overwhelmed.
• Palette-cleanser: a quick way to clear lingering scents between emissions, so one memory doesn't bleed into the next.
• Ambient-scent filter: a way to block unwanted scents from the environment, so users control what they take in (could be considered an additional feature, in the future).

What 83 hours with a strange sense taught me